Inspect PsyCourse data

library(tidyverse)

# remove all objects to start with clean environment
rm(list=ls())

load("W:/group_transbio/data/psycourse/231107_PsyCourse6.0_Pheno_Proposal_69_Schwarz.RData")
data <- prop
rm(prop)

hist_of_differences <- function(data, var1, var2) {
  var1 <- ensym(var1)  
  var2 <- ensym(var2)
  data %>%
    select(!!var1, !!var2) %>%
    filter(complete.cases(.)) %>%
    mutate(neg_change = !!var1 - !!var2) %>% 
    pull(neg_change) %>% 
    hist(main = paste("Histogram of",var1, "-",var2))
}

replace_na_and_integers <- function(x) {
  # Replace NA with "-"
  x[is.na(x)] <- "-"
  # Replace integers with "+"
  x[!is.na(x) & !is.na(as.numeric(x)) & x == as.integer(x)] <- "+"
  return(x)
}


Missing_pattern <- function(data, var1, var2, var3, var4) {
  var1 <- ensym(var1)
  var2 <- ensym(var2)
  var3 <- ensym(var3)
  var4 <- ensym(var4)
  data %>% 
    select(!!var1, !!var2, !!var3, !!var4) %>% 
    mutate_all(replace_na_and_integers)
}

library(dplyr)

# Create a sample dataframe
df <- data.frame(
  var1 = c(1, 2, 3),
  var2 = c(4, 5, 6),
  var3 = c(7, 8, 9)
)

# Vector of variable names to select
variables_to_select <- c("var1", "var3")

# Select variables based on the vector of variable names
selected_df <- df %>%
  select(one_of(variables_to_select))

# Print the selected dataframe
print(selected_df)

var1_vector <- c("v1_panss_sum_pos", "v1_panss_sum_neg", "v1_panss_sum_gen", "v1_panss_sum_tot", "v1_idsc_sum", "v1_ymrs_sum", "v1_gaf",
                 "v1_nrpsy_tmt_A_rt", "v1_nrpsy_tmt_A_err", "v1_nrpsy_tmt_B_rt", "v1_nrpsy_tmt_B_err", "v1_nrpsy_dgt_sp_frw", "v1_nrpsy_dgt_sp_bck", "v1_nrpsy_dg_sym",
                 "v1_nrpsy_mwtb")
var2_vector <- sub("^v1", "v2", var1_vector)
var3_vector <- sub("^v1", "v3", var1_vector)
var4_vector <- sub("^v1", "v4", var1_vector)


results <- 
  data %>% 
  filter(v1_stat == "CLINICAL") %>%
  mutate_all(~replace(., .==-999, NA)) %>% 
  {mapply(hist_of_differences, MoreArgs = list(data = .), var1 = var1_vector, var2 = var4_vector, SIMPLIFY = FALSE)}

Error in `select()`:
! Can't subset columns that don't exist.
✖ Column `v4_nrpsy_mwtb` doesn't exist.
Backtrace:
  1. ... %>% ...
 10. dplyr:::select.data.frame(., !!var1, !!var2)

data %>%
  mutate_all(~replace(., .==-999, NA)) %>% 
  filter(v1_stat == "CLINICAL") %>% 
  select(v1_age, v1_spec_emp, v1_disabl_pens, v1_wrk_abs_pst_5_yrs) %>% 
  filter(v1_age<65) %>% 
  # select(v1_age) %>%
  # pull %>%
  summary()

data.plot <-
data %>%
  filter(v1_stat == "CLINICAL") %>%
  # select(v1_id, matches("panss_sum|bdi2_sum")) %>%
  select(v1_id, matches("bdi2_sum")) %>% 
  rename(id = v1_id) %>% 
  pivot_longer(cols = starts_with("v"),
               names_to = c(".value", "test"),
               names_pattern = "^(v\\d+)_(.*)$",
               values_to = "value") %>% 
  filter(rowSums(is.na(.)) <= 2)

data.plot <-
data %>%
  filter(v1_stat == "CLINICAL", v1_age<25) %>%
  # select(v1_id, matches("panss_sum|bdi2_sum")) %>%
  select(v1_id, matches("bdi2_sum")) %>% 
  rename(id = v1_id) %>% 
  pivot_longer(cols = starts_with("v"),
               names_to = "timepoint",
               values_to = "value")

ggplot(data.plot, aes(x = timepoint, y = value, group = id, color = id)) +
  geom_path(show.legend = F) +
  theme_minimal() +
  labs(x = "Timepoint", y = "Value", title = "Line plot of data for 4 timepoints")